As of September 2025, US law gives no legal privilege protecting users' conversations with AI comparable to doctor or lawyer confidentiality.

Judged as of 2025-09-11. True: legal analyses or reporting show that, at that date, no US federal or state statute and no court ruling had established a legal privilege (comparable to doctor-patient or attorney-client privilege) protecting users' conversations with AI chatbots, so such conversations could be obtained through discovery or subpoena like other records. Mostly true: no general privilege existed, but some narrow existing doctrine could protect certain AI conversations in specific settings, for example work product when AI is used at a lawyer's direction. Mixed: some jurisdiction had enacted or recognised a limited AI-specific confidentiality protection. Mostly false: a significant jurisdiction had a substantial protection of this kind in force. False: a general legal privilege for AI conversations existed in US law at that date.

Set before the research began.

1. E1 Private Thoughts, Public Evidence: AI Chat Conversations ... Tyson & Mendes LLP · 17 February 2026

   “Communications with AI chatbots do not carry attorney-client privilege, psychiatric privilege, or any other type of privilege.”

2. E2 Private Thoughts, Public Evidence: AI Chat Conversations ... Tyson & Mendes LLP · 17 February 2026

   “In a recent 2025 federal court discovery ruling, In re OpenAI, Inc. Copyright Infringement Litigation (S.D.N.Y.), U.S. District Judge Sidney H. Stein compelled OpenAI to produce 20 million de-identified ChatGPT conversation logs to plaintiffs”

3. E3 Your Employees Are Using ChatGPT and Creating ESI – Is It Discoverable? Harris Beach Murtha · 22 September 2025

   “This is critical because it means even chats that are manually deleted by users are being maintained on OpenAI’s servers rather than being automatically deleted after 30 days. As such, those chats are theoretically subject to subpoena.”

4. E4 Bonnie and Claude: The Court Ruling That Could Expose Fraudster’s AI Conversations | Ward and Smith, P.A. Ward and Smith, P.A. · 11 March 2026

   “A New York federal court just issued the first published ruling to tackle head-on whether conversations with a public AI chatbot are protected by attorney–client privilege or the work product doctrine.”

E3, a legal analysis published 11 days after the claim date, says ChatGPT chats kept on OpenAI's servers are "theoretically subject to subpoena." E2 describes a 2025 S.D.N.Y. discovery order that compelled OpenAI to produce ChatGPT conversation logs. E1, a 2026 analysis, states that AI chatbot communications carry no attorney-client, psychiatric or other privilege. E4, a March 2026 article, describes a New York federal ruling it had just reported as the first published ruling to address directly whether such chats are privileged.